Method and Apparatus for Detecting Suspicious Activity Using Video Analysis

1. A method, comprising: providing executable instructions to a processor of a device from a non-transitory computer-readable storage medium causing the processor to perform operations comprising: monitoring a video of a transaction area during a transaction by maintaining a map of a background image for the transaction area and determining based on updates to the background image from the map when objects are present and counting the objects as video item counts from the video; obtaining transaction data for the transaction, wherein the transaction data comprises transaction item counts; and identifying a suspicious activity for the transaction based on a discrepancy between visual information and the video item counts obtained from the video and the transaction data with the transaction item counts.

2. The method of claim 1 , wherein monitoring further includes identifying items for the transaction from the video as a first portion of the visual information.

3. The method of claim 2 , wherein identifying the items further includes maintaining a total count of the items as a second portion of the visual information.

4. The method of claim 3 , wherein maintaining further includes tracking each of the items within regions defined within the transaction area from the video.

5. The method of claim 4 , wherein tracking further includes assigning a region activity for each item within the visual information.

6. The method of claim 5 , wherein obtaining further includes assigning events to each item detected within corresponding regions based on the transaction data.

7. The method of claim 6 , wherein assigning further includes assigning a pattern to each of the items based on the corresponding regions associated with that item and based on the corresponding region's activities assigned to that item.

8. The method of claim 7 , wherein obtaining further includes comparing the corresponding patterns associated with each item against the transaction data.

9. A method comprising: providing executable instructions to a processor of a device from a non-transitory computer-readable storage medium causing the processor to perform operations comprising; tracking items through regions of a transaction area during a transaction from video by analyzing the video based on maps maintained for background images of the regions and determining based on updates to the background images when objects are present and further counting the objects as video item counts for the items; assigning item events to the items based on the regions; assigning a sequence for each item through the regions; defining a pattern to each sequence; obtaining transaction events for the items generated as the items are scanned during the transaction and maintaining transaction item counts for the items; and determining a transaction discrepancy based on the item events, the patterns, the transaction events, the video item counts, and the transaction item counts.

10. The method of claim 9 further comprising assigning the transaction discrepancy to a portion of the video that corresponds to the transaction discrepancy.

11. The method of claim 9 , wherein determining further includes determining the transaction discrepancy when a particular item image for a particular item captured from the video and associated with a particular item event does not correspond to a stored item image associated with a corresponding transaction event.

12. The method of claim 9 , wherein determining further includes determining the transaction discrepancy when a particular item event does not have a corresponding transaction event.

13. The method of claim 9 , wherein determining further includes determining the transaction discrepancy based a particular item pattern being associated with a known discrepancy.

14. The method of claim 9 , wherein determining further includes performing a total item count based on the item events and a total transaction item count based on the transaction events and determining the transaction discrepancy based on the total item count and the total transaction item count being different.

15. The method of claim 9 further comprising, assigning a suspicion level to a segment of the video based on the transaction discrepancy.

16. The method of claim 15 , adjusting the suspicion level assigned to the segment based on an operator identifiers associated with an operator performing the transaction.

17. A system, comprising: a camera configured to provide video of a transaction area for a transaction terminal during a transaction; the transaction terminal comprising a transaction processor and a transaction non-transitory computer-readable storage medium; a server comprising a server processor and a server non-transitory computer-readable storage medium; the transaction non-transitory computer-readable storage medium comprising transaction executable instructions; the server non-transitory computer-readable storage medium comprising server executable instructions; the transaction executable instructions when executed by the transaction processor from the transaction non-transitory computer-readable storage medium cause the transaction processor to perform first operations comprising: providing transaction data as item codes for items scanned or entered at the transaction terminal; and the server executable instructions when executed by the server processor from the server-non transitory computer-readable storage medium cause the server processor to perform second operations comprising: obtaining visual information for the items within the transaction area from the video by analyzing the visual information based on a map maintained for a background image of the transaction area and determining based on updates to the background image when objects are present and further counting the objects as video item counts for the items; obtaining the transaction data from the transaction executable instructions and identifying transaction item counts from the transaction data; and determining a transaction discrepancy based on comparing the visual information with the video item counts against the transaction data with the transaction item counts.

18. The system of claim 17 , wherein the server is further configured to determine the transaction discrepancy when a first item count for the items determined from the visual information does not match a second item count for the items determined from the transaction data.

19. The system of claim 17 , wherein the server is further configured to determine the transaction discrepancy based on a pattern of activity for a particular item within the transaction area determined from the visual information and the transaction data.

20. The system of claim 17 , wherein the server is further configured to determine the transaction discrepancy based on a particular captured item image for a particular item from the video not corresponding to a known item image for a second item identified from the transaction data.